The Federal Trade Commission reports that upwards of 25.6 million Americans fall victim solely to credit fraud, specifically those living in low-income communities. Our team aspires to provide a solution and positively impact the 40 million low-income Americans currently at risk to credit and other investment-related fraud. Benford's Law states that the first digits of a natural set of numbers follow a logarithmic distribution. In other words, for a set of natural numbers, the first digit is "1" around 30% of the time, "2" around 20% of the time, and so on, until it is "9" around only 4% of the time. The only condition for distributions to follow Benford's Law is for them to be natural distributions. Therefore, Benford's Law can be used to differentiate between natural and fabricated sets of data.
What it does
Typically, only mathematicians and statisticians are believed to have enough expertise to perform Benford's Law calculations. However, that is not true and because it has been proven mathematically, can be applied with certainty to a vast majority of data in our daily lives. Our program is an applet using the R-programming language that allows a user to input data and determines whether the data is natural or fabricated using Benford's Law. In making it easier to determine the fraudulence for a set of data, our application has the ability to impact tens of millions of lives every day from becoming prone to fraud.
How we built it
Challenges we ran into
It was extremely difficult to debug our R applet at certain times because of the limited documentation and knowledge in the language, especially since it is our first time creating an applet using this programming language. Additionally, it was difficult to program the mathematical aspect of Benford's Law and knit the documents together using RHTML.
Accomplishments that we're proud of
We are proud of being able to use several real-life sets of data and to accurately determine whether the data has been fabricated or not such as world populations and data from recent bank scandals. Additionally, it is our first time using R to program such a nuanced mathematical concept and retrieve the output of an applet.
What we learned
We learned a lot about the R-programming language and how R and HTML can communicate through RHTML. Additionally, we learned a lot about tax fraud in our current world and how women in developing countries and those living in low-income households are particularly targeted and negatively impacted by credit fraud.
What's next for PFF: Preventing Future Fraud
In the future, we hope to expand the use of this program and improve the user interface so it becomes even easier for those in low-income communities and developing nations to use for a practical application.